search

michigan-traffic-lab / Dense-Deep-Reinforcement-Learning

This repo contains the code for paper "Dense reinforcement learning for safety validation of autonomous vehicles"
Other
317 stars 49 forks source link

Code for D2RL algorithm #1

Closed zincopper closed 1 year ago

zincopper commented 1 year ago

Would you happen to provide the source code for the D2RL algorithm publicly? I have only been able to find the trained parameters in checkpoints. It would be greatly appreciated if the source code for the algorithm could be provided. Thank you.

Wi11ione commented 1 year ago

Currently there is no "DRL_gym_ENV_offline" method provided for generating training environments in the training part of D2RL. If you could provide the code for the D2RL training environment, I would greatly appreciate it!

Zoe-Wan commented 1 year ago

Hello! Your work is very interesting and I wonder if you can publish the detail of the training part? I 'm reproducing your work and I really need it!!

miracelplus commented 1 year ago

Hi all,

The code for D2RL training algorithm has been uploaded to this repo. Please refer to README_D2RL_Train.md for detailed guidance.

Best,

Haowei

hugeshark7 commented 12 months ago

Hi Haowei, when i follow the instruction runing the D2RL_train.py, it shows no module of RLlib.agents. I do see the initial file, but i have no idea how to register an agent. By the way, can I modify the maps file to train other agents using this D2RL_train.py. I would really appreciate your reply!